b ‘

Reprinted for private circulation from. el 4 à. DIE ga
| Vol. XIV, No. 11, March 1917 ME

Su ;

THE BRIEF-NARRATIVE ART OF THEOPHILE GAUTIER

When Maxime Du Camp affirms! that Gautier is less of a roman-
cier than a conteur, he is attempting to distinguish between these as
between invention and imagination, arguing that whereas a roman is
composed objectively, upon a deliberate plan, a conte or a nouvelle
is subjective and spontaneous. This distinction, carried to its logical
consequences, means that in novels the writer guides the narrative, in
brief tales the narrative guides the writer, a reduction to the absurd
even if limited to Gautier. For the structural unity of la Morte
amoureuse is as voluntary as that of le Capitaine Fracasse, and vastly
superior to that of such novels as Partie carrée. Du Camp is mani-
festly correct in assuming that many of Gautier’s briefer tales are
the result of musing over adventures, generally erotic, of which the
author imagines himself the hero, and there is a degree of reason in
his remark that “c’est parce qu’elles ont été un épisode de sa vie
intellectuelle que ses nouvelles sont simples, presque sans incidents,
émues néanmoins et communiquant l’émotion dont elles palpitent.’’?
But this subjectivity is not in itself an adequate explanation of the
peculiar singleness of effect of some of the stories, and the question
remains by what narrative methods Gautier achieved this.

The ultimate solution should include an appraisal of Gautier’s
importance in the development of brief fiction in France, but this
can be attempted only after such a survey of the entire field as no one
has yet made. My immediate purpose is to examine Gautier’s tales
for themselves and to discover whether he evolved a type or types of
any narrative distinction.

In a group of the earlier stories, notably in la Morte amoureuse
(1836), his methods are comparable to Poe’s, although there is no
likelihood that at this period Gautier was acquainted with the writ-
ings of the American.

Whoever analyzes la Morte amoureuse will perceive that it fulfils
the short-story requirements later expounded by Poe and so diligently

1 Théophile Gautier (Paris, 1890), pp. 147-52.

2 Ibid., p. 149. :
647] 135 [MopERN PHILoLoGy, March, 1917

136 Horatio E. SMITH

studied by his followers and his critics.! Gautier chooses for a
theme the liaison of a priest and a female vampire and sets out, as
the short-story adepts now do, “to produce a single narrative effect
with the greatest economy of means that is consistent with the utmost
emphasis.”? His devices for procuring complete harmony are legion.
The point of view is constant, for the whole experience is related in
the first person by the victim. The initial sentences tend to the
outbringing of the preconceived effect, to sounding that note of the
uncanny which is constantly to be maintained: “‘ Vous me demandez,
frére, si j’ai aimé; oui. C’est une histoire singuliére et terrible, et,
quoique j’aie soixante-six ans, j’ose à peine remuer le cendre de ce
souvenir. * The entire first paragraph, summing up the indelible
impression made upon the priest, is in the same key, and throughout
the tale the note rings out, ever lugubrious.4 Except for supernumer-
aries entirely inconspicuous, there are only three characters, the vam-
pire, the priest Romuauld, and a brother-priest whose attempts to
offer guidance form an essential part of the action. In the matter of
descriptions, tabooed in the short-story as foreign to its essence,
Gautier, who is notoriously fond of the picturesque, has restrained
himself to an unusual degree. The events are simple and decisive,
and at turning-points in the story attention is fastened upon the
stage of development reached by a terse summarizing phrase.’
Suspense is brought skilfully to a head by the cumulation of mani-
festations made to the priest by Clarimonde. When Romuauld is
summoned to the unfamiliar chateau, the promptness with which he
realizes that he is to see his lady is not only natural but a factor—
one of a score—in securing swift progress. The ending is as direct

1 For an abridged short-story bibliography, cf. my article on ‘‘ Balzac and the Short-
Story,’’ Modern Philology, XII, 71, note 3. The famous statement of Poe, part of which
is quoted in that note, was written in 1842. The short-story character of la Morte
amoureuse has already been pointed out by Professor Baldwin, American Short Stories
(New York, 1909), Introduction, p.33. In the present article I shall use the term ‘‘ short-
story’’ only in the restricted American sense.

2 Hamilton, Materials and Methods of Fiction (New York, 1908), p. 173.
3 Nouvelles (Paris, 1871), p. 261.

4 Observe the repetition of the idea of a nightmare (ibid., pp. 266, 275, 289); of the
apostrophe of Clarimonde: ‘ Malheureux! qu’as-tu fait?’’ (pp. 267, 268, 295); of the
motif of a single fatal glance: ‘‘un seul regard . . . jeté sur une femme”’ (p. 261), ‘pour
avoir levé une seule fois le regard sur une femme’’ (p. 274), ‘‘ne regardez jamais une
femme . . . il suffit d’une minute”’ (p. 295).

5 Pp. 263, 268, 288.
648

Tue Brier-NARRATIVE ART OF THEOPHILE GAUTIER 137

as the beginning: the liaison is broken, Clarimonde bids the priest
farewell, ‘‘elle se dissipa dans l’air . . . et je ne la revis plus.”
There follows a paragraph of a few lines which sound for the last time
the note of the awfulness of the experience and is simply long enough
to save the termination from being brusque, to put the needed period
to the whole.

La Morte amoureuse was prepared for Balzac’s Chronique de
Paris,? and Balzac himself had written a little earlier* several tales
of salient short-story characteristics, yet neither he nor Gautier in
any way suggest that they were aware of the temporary similarity
of their narrative processes. A similarity there is, however, and not
in the case of la Morte amoureuse alone; other stories written by
Gautier within a few years preceding 1836 approach the same stand-
ard and may be considered tentative efforts at the form finally
achieved in 1836.

The first of these is la Cafetiére (1831). A young man, Théodore,
staying at a country house, has a singular adventure during the night,
and, like the priest Romuauld, he tells of the experience himself.
Objects in his room come to life, a porcelain coffee-pot thumps its
way from table to hearth, portraits become animated and dance,
and the youth, perceiving a charming girl without a partner, joins her.
At dawn the spell is broken, the girl falls to the floor, and Théodore,
rushing to pick her up, finds only the pieces of the coffee-pot. The
exposition is of the briefest, the narrative progress is swift, the
attention is focused, first upon the weird group of figures in the room
and then, sharply, upon the principal figure, Angéla.4 A single tone,
suggesting the weird and the tragic, prevails as in la Morte amoureuse,
and is last sounded in the final words of the tale when the hero
expresses his—somewhat callow—despair with the remark: “Je
venais de comprendre qu’il n’y avait plus pour moi de bonheur sur
la terre!” In a tentative version of the first part of the story,

1P. 295.

2 Lovenjoul, Histoire des œuvres de Théophile Gautier (Paris, 1887), No. 130. The
story was divided between two issues of the Chronique de Paris. Apparently Gautier had
no thought of the advantage which, according to Poe, results from a single, uninterrupted
presentation.

3 Cf. p. 136, note 1.

4 In 1852 the story was published under the title Angéla (Lovenjoul, op. cit., No. 45).

5 Les Jeunes-France (Paris: Charpentier), p. 261. In this edition the story is dated,

rroneously, 1833.

649

138 Horatio E. SMITH

reproduced from the manuscript by Lovenjoul,! there are copious
descriptions of no narrative value, and a more cumbersome exposition,
and in the fact that this version was rejected there may be a sign of
a conscious move to attain a higher degree of unity.

Onuphrius (1832) consists of a series of adventures of a young man
who gradually becomes insane. The events are related with remark-
able imaginative power and are somewhat varied, too varied no doubt
to admit classification of the story with the Poe type. But the basic
unity is complete. Onuphrius has something in common with le
Horla by Maupassant, a tale worthy of Poe at his best, resembling it
closely enough to prove short-story characteristics, differing suffi-
ciently to indicate short-story defects. The experiences in le Horla
are as diversified, but there results a high degree of unity from the
fact that the illusion of the patient is a single one, while that of
Onuphrius, except for the recurring hallucination of the demon with
the ruby on his finger, varies. The climacteric development in le
Horla is superior, and the attention is more completely focused on the
man and his obsession. Le Horla is told in the first person, whereas
with Gautier, although the point of view is constant, the third per-
son is used, with consequent diminution, it may be argued, of in-
tensity.”

Three other products of the years immediately preceding la Morte
amoureuse merit a word. Two of them are hardly more than anec-
dotes. The first, published by Lovenjoul,*? who without vouching
altogether for its authenticity thinks it may be the first piece of
fiction ever composed by Gautier (March 24, 1831), is the account of
the experience of a young Frenchman in Egypt, compelled to seek the
hospitality of a Bedouin camp and discovering that he has eaten of
a gazelle roasted over blazing mummies. The next is the story, cover-
ing some eight pages in Sous la Table (1833) ,* concerning a youth and
a grisette. The episode is too trivial to warrant serious consideration,
yet, its essential narrative structure alone considered—and it is

1 Louvenjoul, op. cit., I, pp. 16-20.

2 The ending of the first edition, quoted by Lovenjoul, op. cit., No. 56, is more abrupt
and less artistic than that of the Jeunes-France volume.

3 Op. cit., I, pp. 8-11.
4 Les Jeunes-France, pp. 11-18.

— eo ee ee ee ee ee, ee

Tue Brier-NARRATIVE ART OF THEOPHILE GAUTIER 139

sufficiently developed to be said to have a structure—the singleness
of effect is achieved by a process comparable to Poe’s. In the third,
Omphale (1834), the narrative current is swift and steady. Omphale
is a lady portrayed on an old piece of tapestry in a young man’s room.
She comes to life and loves him, but the affair is interrupted by the
young man’s uncle, who removes the tapestry. The account of the
subsequent effort of the youth to regain the tapestry is not, however,
acceptable from the short-story point of view, and, a more vital
matter, the chief interest is in the picturesque setting, as the sub-
title, histoire rococo, suggests.

These of course do not constitute all of the brief tales written
by Gautier between 1831, at the beginning of his career, and 1836,
but the others are nondescript, and some of them, such as le Bol de
punch, a picture of a revel, and Daniel Jovard, a character sketch,
are narrative only in name. The point is that during this period
Gautier write one genuine short-story and other pieces approaching
the type. As has been suggested, it is not safe to posit in this group
any influence of Poe, who was only beginning to write at this period
and who was not known in France, at least in translation, until the
forties.! The truth seems to be that Gautier is one of several authors
who were at that time creating in France, independent of foreign
models, a type the identity of which with Poe’s is at once fortuitous
and complete?—and which I label short-story for want of authority
to call it anything more distinctive. But after 1836 Gautier writes
no more narratives of this form; his methods continue to develop,
and, oddly enough, by the time it may be assumed that Gautier
was well acquainted with Poe, he is producing stories of a stamp
distinct from the American’s.

The prime difference is in degree of compression. Superlative
concision, the sine qua non of the Poe type, is no longer achieved or

1 According to Retinger, le Conte fantastique dans le Romantisme francais (Paris,
1909), p. 33, note, The Murders in the rue Morgue appeared anonymously as early as 1841.

Cf. Lauvriére, Poe (Paris, 1904), p. 644, note 2; p. 276; Morris, Cooper et Poe d’après la
critique française du diz-neuviéme siècle (Paris, 1912), pp. 80, 203.

2 Professor Baldwin, op. cit., Introduction, p. 33, pointing to the dates of la Morte
amoureuse (1836) and of Berenice (1835), says: ‘‘ Remarkable as is the coincident appear-
ance in Paris and in Richmond of a new literary form, it remains a coincidence.’’ Cf.
Canby, Study of the Short Story (New York, 1913), p. 45; ‘‘ The work of their [the Ameri-
can writers’] French contemporaries . . . . represents a parallel, not a derivative move-
ment.’’

651

140 Horatio E. SMITH

sought. What this means in actual practice will become evident
upon inspection of the significant examples.

The tendency is already manifest in Une Nuit de Cléopâtre (1838).
A young Egyptian, so enamored of Cleopatra that he dares penetrate
to her bath, is there discovered by the capricious queen, who accepts
him as her lover with the condition that the next morning he shall
die. At daybreak he pays the penalty. This would be an admirable
nucleus for a short-story, and it is not difficult to imagine how the
theme would be treated. The bounds of the above summary would
be respected, the action would hardly begin before the capture of the
youth at the bath, and would cease with the administration of the
poison. In the latter case Gautier has, unconsciously, met the short-
story requirement; his termination, according to that convention, is
impeccable, possessing even that unexpected twist so sought after
by the American twentieth-century writer, often with a resultant
artificiality—though the turn which Gautier gives the narrative
development is altogether probable as well as interesting: at the
crucial moment Cleopatra wavers and considers preventing her lover
from drinking the potion, when the unlooked-for announcement of
the arrival of Mark Antony precipitates the catastrophe. The story
at this point moves as rapidly as Poe’s followers say it must. But
here only. Gautier does not introduce the episode of the bath until
he reaches the fifth of the six chapters. His preparation is leisurely.
In the first chapter we learn that Cleopatra is bored, in the second
this theme is elaborated and we have a first glimpse of the lover, in
the third the lover becomes active to the extent of sending Cleo-
patra a message by an arrow, in the fourth we learn that Cleopatra is
no longer bored and that the youth contemplates further activity.
This easy-going preliminary, the frequent changes in viewpoint and
in place, and above all the copious descriptions, are quite contrary to
short-story practice.

In the case of le Roi Candaule (1844), Gautier appears to have
been of two minds. He was evidently tempted at the outset to
yield to the charm of the picturesque, and then became interested in
the development of the action and gave over the static for the kinetic.
An admirable narrative is developed about a central point, the in-
spection of the queen’s beauty. The author seems to have taken,

652

THe Brier-NARRATIVE ART OF THÉOPHILE GAUTIER 141

for structural purposes, the point of view of Gyges.! At the begin-
ning it is not apparent that the young captain is to play an important
role, for the first chapter consists almost entirely of an elaborate
description of the marriage of King Candaules and Nyssia, but, as the
story moves on, more and more light is centered upon Gyges, the
descriptions which, from the point of view of narration, obstruct,
become less and less frequent, and after an extremely leisurely start
the action becomes rapid. The preparation for the climax is adroit.
It is necessary to explain concisely, when the crisis approaches, how
it happens that Candaules wishes to have another man behold his
wife’s charms, and this is the more easily done since it has already
been casually developed that he is a person of uncommon stamp and
especially that his appreciation of beauty may be that of the artist
as much as that of the lover. The keenness of vision which enables
Nyssia to discover the eyes of Gyges as he looks into the chamber
from his hiding-place is suggested early in the story. Surely such
arrangement of details demonstrates what Maxime Du Camp would
deny, the exercise of inventive powers by the conteur? The termina-
tion of the story, after cleverly maintained suspense, is swift; climax
and end are almost simultaneous. The murder done, the problem
solved, a brief final paragraph gives a last glance at the central theme:
“le nouveau roi se maintint sur le trône de Lydie, qu’il occupa
pendant de longues années, vécut heureux et ne fit voir sa femme à
personne, sachant trop ce qu’il en cofitait.’”

Somewhat parallel in structure is Arria Marcella (1852).4 A
young Frenchman, at a museum in Naples, is moved by a relic of

1 The following is a summary of the chapters from this angle: (i) it transpires that
Gyges, by a rare chance, has seen the queen unveiled before her marriage; (ii) Gyges is
the chosen confidant of the king as regards her beauty; (iii) Gyges sees her in the cham-
ber; (iv) he is summoned to the queen to avenge her; (v) he wins her.

Le Roi Candaule is clearly enough an adaptation of Herodotus’ story of Gyges
(i. 8-12), itself an excellent example of narrative economy; the principal incidents in
Gautier are those related by the Greek, and several items of the conversation between
Gyges and Candaules and of that between Gyges and the queen are almost exactly repro-
duced. But Gautier’s elaboration of the theme remains his own.

2 Cf. Taylor, ‘‘The Short Story in France, 1800-1900,”’ Edinburgh Review, July, 1913,
p. 144: ‘‘Mérimée, Gautier, Flaubert, exercised, however covertly, an art of composition;
they disposed their incidents in due order of sequence; they arranged their figures with
an aesthetic sense of perspective, prepared and suspended their crises, and held the
balance of accent and emphasis.”’

* 3 Nouvelles, p. 419.

4 Lafcadio Hearn (One of Cleopatra’s Nights, etc. [New York: Brentano, 1906],

pp. 385-88) thinks that Gautier may have found the inspiration for Arria Marcella in an

653

142 Horatio E. SMITH

Pompeii, a mass of ashes which has hardened about the form of a
beautiful woman perishing in the destruction of the city. With
imagination stirred by a daylight inspection of the ruins he returns
there the same night, finds the conditions of eighteen hundred years
ago restored, encounters the lady in question, and is welcomed by
her, when her Christian father steps in, berates his daughter, and
destroys the hallucination.! Arria Marcella is composed with no
regard for the short-story principle that ‘‘the narrative must move,
move, move furiously each action and every speech pointing directly
toward the unknown climax,’ yet it is carefully organized in the
interest of unity, with graduated and skilful focusing of the attention
upon the heroine and artful alternation of suspense and swift ad-
vance. The exposition and the action proper are amalgamated
by the author’s conception that Arria is already acquainted with the
passion of the young man and knows, by a weird interchange of the
centuries, that he has admired her in the museum.’

Common to these later stories is a feature already observed in
la Morte amoureuse, the deliberate insertion of forward-pointing
remarks which serve to emphasize the central theme. The ultimate
and sinister achievement of Gyges is already faintly suggested, at
the very outset of le Roi Candaule, in the musings of the young cap-
tain: ‘il songeait aux enivrements de la toute-puissance, au bonheur
. . . de poser le diadéme sur la tête de la plus belle,” “en effet,
c'était bien sur ce front [de Nyssia] qu’il eût voulu poser le diadéme,”’
“amour qu’il éprouvait pour Nyssia lui causait une secrète terreur.’”#
The strange adventure of the hero of Arria Marcella is foreshadowed
in the searching glance of his companion, and in his own first impres-
sions: “Octavien . . . semblait plus touché que ses insouciants
compagnons du sort de ces trépassés de deux mille ans,” ‘Les
phrases banales du guide causèrent une vive émotion à Octavien.’’5

old Greek ghost story, and quotes Michelet’s version of this. In any case it is clear that
Gautier’s narrative methods are his own.

1 Observe how frequently the basic plot of a liaison interrupted by a third person
recurs (Omphale, la Morte amoureuse, le Pied de momie, even, in a sense, Une Nuit de
Cléopâtre and le Roi Candaule).

2 Canby, ‘‘Free Fiction,’’ Atlantic Monthly, July, 1915, p. 61. Professor Canby
urges his compatriots not to be fettered by such rules.

3 The period to the story, the final sentence, deftly enhances the unity.

4 Nouvelles, pp. 364, 365, 365, respectively,

5 Romans et Contes (Paris: Charpentier), pp. 272, 280, 282, respectively.

654

Tue Brier-NARRATIVE ART OF THEOPHILE GAUTIER 143

Occasional emphatic sentences, such as the statement in Une Nuit de
Cléopâtre, “il s’était juré . . . qu’il serait l’amant de Cléopâtre

. ne fût-ce qu’une nuit . . . dût-il lui en coûter son corps
et son 4me,’’! are at once a prophecy and a summary, and the progress
of a story is often epitomized, with a resultant and deliberately
attained narrative unity, as in Gyges’ résumé of the course of events:
“Un hasard lui avait fait connaître sa beauté murée à tous les yeux,
entre tant de princes et de satrapes elle avait épousé précisément
Candaule, le roi qu’il servait, et, par un caprice étrange qu’il ne
pouvait s’empécher de trouver presque fatal, ce roi venait faire, à
lui Gygés, des confidences sur cette créature mystérieuse que per-
sonne n’approchait. .. .’”

These features become less conspicuous in three narratives of later
date, much longer than those so far under discussion and yet pos-
sessed of a similar unity distinct from that of the novel. The evolu-
tion appears to be steady, from a narrative organization strictly
limited, with unity consciously emphasized, to one of broader scope.

The theme of Avatar (1856) is a reincarnation and its conse-
quences. Octave de Saville almost despairs at the hopelessness of his
love for the Countess Labinski, who adores her husband. Thanks
to the occult science of an uncanny physician, Octave’s soul is
transferred to the husband’s body, and the husband’s soul is incor-
porated in his. But the lover finds success as impossible as before,
and the souls at his behest are being retransferred when his flies off
into space, and the doctor occupies his body. The progress of the
story is as steady and unhurried as that of Arria Marcella, but the
attention is not equally concentrated. In Arria Marcella the single
interest is in the relations of the youth and the girl of Pompeii;
in Avatar the interest is for a considerable period in the relations of
Octave and the lady, but later, when this situation is solved, much
space is devoted to the resultant fate of the young man. And since,
in the nature of the situation, the reader would be almost as interested
in the effects of the reincarnation on the husband as in its effects on
the lover, there is a considerable discussion of this. The remark of
Gautier, summing up the case for Octave, ‘Ame obscurément
sublime, il ne savait qu’aimer et mourir, * may be regarded as the

1 Nouvelles, p. 345. 2 Ibid., p. 387. 3 Romans et Contes, p. 126.
655

144 Horatio E. SMITH

nucleus of the tale, yet when the interest turns away from Octave
the author does not hold it in check, he does not impose an adventi-
tious unity. The theme of Arria calls for development within a
small circle, that of Avatar occupies a relatively large one.!

Avatar is promptly followed by another story of the same order,
Jettatura (1856). Here Gautier unfolds the consequences of the
Neapolitan superstition of the evil eye when it fastens itself upon a
young foreigner in love with a girl threatened with tuberculosis. At
every point, in composing the tale, he has kept this situation in mind.
His methods as usual are leisurely, but the narrative is held true to its
course. The technique is remote from that of the short-story.
The beginning, where, before the hero is introduced, there are five
pages of easy-going description of the approach to Naples and some
humorous animadversions on British travelers, recalls the lack of
restraint of the novel? Yet there is a deliberate and successful
attempt to secure singleness of effect. Step by step the theme of the
evil eye of Paul d’Aspremont is developed. The thought that there
is a peculiar and pernicious power in his gaze is kept constantly before
the reader. At first there are only mild suggestions. Then dis-
asters which take place in the presence of Paul and which, the reader
is gradually persuaded, may be due to the peculiar power of his eye,
accumulate. At length, at a point where his gaze seems to produce
catastrophes in rapid succession, Paul first hears the epithet jettatore!
His concern over his strange situation and especially over its possible
consequences for Alicia grows rapidly until the climax. Preparation
for the one critical situation and the development of that situation,
or in other words a climacteric elaboration of the power of Paul’s
eye and then an account of the result when this power is concentrated

upon Alicia, such is Gautier’s method.ÿ

1 The twelve sections of the story, as it appears in book form, are presumably the
twelve units of its serial publication (cf. Lovenjoul, op. cit., No. 1400). Here, as well as
elsewhere, Gautier deals successsfully with the exigencies of the feuilleton system; his
division into short chapters is highly logical and his final sentences sometimes epitomize
with the succinctness of the short-story termination, sometimes point forward to the
next stage of the tale in a way that welds the sections together.

2 There are other suggestions of novel technique. Cf. p. 161, where Gautier returns
to the exposition by relating events that antedate the action of the story. Such retro-
gression is rare in tales where a high degree of unity is the desideratum (although it is
found in Stevenson’s Dr. Jekyll and Mr. Hyde). Cf. pp. 252-55, where the author intro-
duces supernumerary characters of somewhat meager interest as exemplifying the idiosyn-
crasies of the British and of almost no usefulness in advancing the narrative.

3 Gautier started to cast Jettatura in verse, and several fragments of his work exist
(ct. Lovenjoul, op. cit., No. 2306). The beginning of the narrative is more direct, the tone
more sinister, than in the prose form. 656

Tue Brier-NARRATIVE ART OF THEOPHILE GAUTIER 145

Spirite (1865), one of his last pieces of fiction, closes the series
of distinctive tales I am here considering. It is of particular interest
in that it was written under the influence of Poe,! with whose work
the author was now thoroughly familiar,? and yet, as regards struc-
ture, is closer to the Avatar type than to la Morte amoureuse.

Once more, maximum compression is not sought. The essential
theme, the relations of Guy de Malivert and the spirit of a woman
who had already loved him in the flesh, without his knowing it, is
given due prominence, yet Gautier does not refrain from dealing with
the immediate ramifications. In addition to Malivert and Spirite
two characters play rôles of some importance, Madame d’Ymber-
court, the flesh-and-blood rival, and the Baron de Feroé, a mystic who
is able to explain to Guy the manifestations of his intangible mistress.
Both are required in Gautier’s manipulation of the subject, the first
as a foil, the second as an interpreter. The scene is allowed to shift,
the point of view changes. The fourteenth chapter, wherein the
main issue waits while the jealousy of Madame d’Ymbercourt is
healed by Spirite, would be considered by Poe an unpardonable
interpolation.2 Even from a more liberal point of view certain
features may be considered to clog the narrative, for example the
long description by Spirite, who is recounting her mortal life, of the
ball where she did not dance with Malivert.4 On the other hand, it
is clear that Gautier felt the need of a certain degree of narrative con-
cision: when the situation becomes critical the story is made to
move rapidly, the writer comes promptly to the essential,° and once
he remarks: “Il est inutile de décrire avec détail les impressions de
voyage de Malivert; ce serait sortir du cadre de ce récit.’’®

1 Cf. Retinger, op. cit., pp. 61, 77. Gautier mentions Poe in the course of the story
(pp. 4, 33-34).

2 I find no reference by Gautier to Poe earlier than 1858 (cf. Portraits contemporains
[Paris, 1898], p. 52). From this time forward allusions are frequent, as if Gautier had
suddenly become aware of the American, perhaps through Baudelaire’s translations (1856).
In his article on Baudelaire, Gautier quotes, without comment, a remark by Poe which
contains in essence the latter’s short-story theory (Portraits et souvenirs littéraires [Paris,
1892], p. 191).

3 Furthermore, Gautier has not, in the Poe manner, ‘‘deliberately preconceived”’
the whole story, to judge from the following (Vacances du Lundi [Paris, 1907], p. 75):
“nous voilà installé au sein d’un doux et charmant loisir, cherchant sous les grands
marronniers la fin de Spirite.’’

4 Perhaps the explanation at the close of this description may be considered faintly
apologetic (Spirite [Paris, 1907], p. 134).

5 Cf. ibid., p. 138.
“ Tbid., p. 218. Observe, however, that a page of description follows.
657

146 Horatio E. SMITH

This note of conscious construction is rare in Gautier, either in
his stories or in his critical writing. If we now examine the latter,
it will become evident that he can hardly be said to have, in the case
of fiction, any doctrine.

Even as regards terminology he is careless. He called Spirite a
nouvelle. Avatar and Jettatura first appeared as contes,! but they
are the longest stories, and stand at the head, in the subsequent col-
lection entitled Romans et Contes (1863). In his autobiography,
referring to his activities in the field of fiction, he writes: “Sans
être romancier de profession je n’en ai pas moins bâclé, en mettant
à part les nouvelles, une douzaine de romans: les Jeunes-France,
Mademoiselle de Maupin, Fortunio, les Roués innocents, Militona, la
Belle Jenny, Jean et Jeannette, Avatar, Jettatura, le Roman de la
momie, Spirite, le Capitaine Fracasse.”” Here is a list which includes
novels, tales that approach the short-story type, and tales that bear
another stamp; they have in common merely the fact that they are
relatively long. There is apparently no distinction between roman
and nouvelle, unless it be one of length, in the following remark: “Il
y a douze ou quinze ans, M. Jules Janin publia, dans la Revue des
Deux Mondes, une nouvelle d’une centaine de pages—on n’avait pas
encore inventé, en ce temps-là, les romans qui n’en finissent pas—
une nouvelle, disons-nous, intitulée le Pzédestal.”* And certainly
little difference between nouvelle and conte is recognized by Gautier
when he uses the two words in the same sentence of the same story.*

A specific problem of narration is discussed by him to some pur-
pose in his article on Karr. Of the digressions in tie novels of the
latter he writes:

+ Lovenjoul, op. cit., Nos. 1400, 1421. 2 Portraits contemporains, D. 13.

3 Histoire de l'art dramatique en France (Paris, 1859), VI,5. Le Piédestal was pub-
lished in 1832, notin the Revue des Deux Mondes but in the Revue de Paris. Janin,inthe
course of the story, makes a significant distinction between nouvelle and roman
(XLIII, 103): ‘‘ Ici si je faisais un roman et non pas une histoire, j'aurais un bien beau
sujet de développements de mœurs. J’arrangerais à loisir mon récit, le conduisant en
habile écuyer à travers toutes les difficultés du terrain, changeant souvent ma voie.

‘“ Mais il n’en est pas de la nouvelle comme du roman. La nouvelle, c'est une course
au clocher. ... On va toujours au galop, on ne connaît pas d'obstacles; on traverse le
buisson d’épines, on franchit le fossé, on brise le mur, on se brise les os, on va tant que
va son histoire.’ In performance, Janin proves somewhat irresponsible. His composi-
tion is lax. But the metaphor brings out effectively the essential directness of the nou-
velle type, and, although Gautier passes it over in silence, suggests the economy of the
short-story art at which he himself was trying his hand in 1832.

4 Histoire de l'art dramatique, I, 254, 299; Lovenjoul, op. cit., No. 830.
658

Tue Brier-NARRATIVE ART OF THÉOPHILE GAUTIER 147

Un mot fait éclore un chapitre, et malgré toutes leurs folles brindilles
éparpillées à droite et à gauche, ces digressions n’en tiennent pas moins à la
tige commune par des filaments et des nervures invisibles. Relevez le
feuillage de la main, et vous verrez la branche qui s'attache solidement au
trone; toute action, si elle a réellement une portée philosophique, fait lever
une moisson de pensées sous lesquelles il lui arrive quelquefois de disparaître
comme la terre aride du sillon sous le manteau d’or des épis.—Lequel vaut
mieux de l’épi ou du sillon, de la feuille ou de la branche ?!

No doubt the principle was applied by Gautier himself, although
with modifications and a limited portée philosophique, and perhaps
the remark is the inspiration of Du Camp when he says of Gautier’s
workmanship: “Le sujet . . . est toujours d’une extrême simpli-
cité, mais l'écrivain a su le parer et l’envelopper, parfois jusqu’à
le faire disparaître, d’une forme élégante et touffue.”’? Similar
praise of inherent unity, deserving of quotation in a discussion of
Gautier’s own procedure, is found in his encomium of Fenimore
Cooper:

Les plus beaux romans de Cooper sont composés avec des éléments d’une
simplicité extrême. .. . Les personnages n'apparaissent que comme des
points blanes et rouges sur le fond d'outre-mer des lointains, ou sur le vert
sombre et dur des ébéniers centenaires. Cependant, malgré leur petitesse
relative, par leur énergie et leur résolution, ils dominent cette gigantesque
nature, et c’est là la source de l'intérêt sublime et profond qui s'attache au
Dernier des Mohicans, à la Prairie. L’orgueil humain est intimement flatté
de cette victoire et s’en réjouit par esprit de corps. Cette disposition rendait
Fenimore Cooper plus propre que tout autre à réussir dans le roman maritime.
_. . L'idée qui éclate à chaque page est celle exprimée par le proverbe bre-
ton: “Ma barque est si petite et la mer est si grande!” De là vient tout
Vintérét.?

But these acute bits of criticism stand almost alone. An able
practitioner, Gautier, as I have said, was not given to theorizing
about fiction.

It is therefore upon the examination of the structure of his stories
that conclusions must largely be based. Two types appear to have
been evolved, a form exceedingly compact, best represented by la
Morte amoureuse, and another less restrained but still highly unified,

1 Portraits contemporains, p. 17.

2 Du Camp, op. cit., p. 151.

3 Souvenirs de théâtre, d'art et de critique (Paris, 1904), pp. 21-22. This is part of a
page of criticism of Cooper which occurs in an article on Eugène Sue first published in
the Chronique de Paris in 1836. It is not listed by Professor Morris, op. cit.

659

148 Horatio E. SMirx

of which Arria Marcella and Jettatura are characteristic examples.
The one develops out of the other; the first group is the result of
an early tendency culminating in 1836, the second includes some of
the best work of the author’s maturity.!

Needless to say, many of his narratives do not yield to this
classification. Some are not primarily narrative but descriptive ;?
some are genuine novels;? some, within a narrower compass, retain
the characteristic structural complexity of the novel.4 Even such a
tale as le Chevalier double, in spite of its strongly emphasized central
theme, might be considered a novel in parvo, were it not unreason-
able to apply modern narrative standards to a story which aims at
and achieves the characteristics of an ancient legend.’ La Toison
d’or is an intermediate form. The story is an organic whole; the
central theme is the quest of Tiburce, a youth who, having admired
womankind only from the point of view of a connoisseur of the fine
arts, only in painting and sculpture, determines to seek a flesh-and-
blood mistress and, inspired by Rubens, goes to Flanders. At the
beginning the interest is fastened upon the question whether Tiburce
will find a blond maiden, then it is deflected to the infatuation of
Tiburce for Rubens’ Magdalene, and ultimately to that development
of the character of Gretchen which results in the final solution, with
the affections of Tiburce satisfied and his power as an artist revealed.
In other words, the attention is not focused. It is not difficult to
imagine the expansion of la Toison d’or into a full-sized novel, where
the character development, no longer compressed into a few lines,
would be vastly more convincing; on the other hand, the story
could hardly be given a unity comparable to that of le Rot Candaule
or of Avatar without fundamental revision.

1 In addition to the stories discussed, la Chaîne d'or (1837) and especially la Mille et
deuxième nuit (1842) belong to the second group.

2 Le Pied de momie, le Pavillon sur l'eau, even Fortunio, which is a cross-section of a
peculiarly sumptuous and voluptuous life with the central figure hardly a man but a
personification of a picturesque lavishness.

3 Le Capitaine Fracasse, le Roman de la momie, etc. The plot of the second is rela-
tively simple. Cf. the simplicity of plot of la Croiz de Berny, written by Gautier in
collaboration with Mme de Girardin, Sandeau, and Méry.

4 Militona, le Berger, etc. So also les Roués innocents, and Jean et Jeannette, although
in each case the elaborations develop from a single nucleus.

5 Other legends and fairy-stories by Gautier approach more closely, in narrative
simplicity and concentration of attention, the types under discussion. Cf. the exquisite
Enfant aux souliers de pain and Nid de rossignols, and the mediocre Oreiller d'une jeune
fille.

660

Tue BRIEF-NARRATIVE ART OF THEOPHILE GAUTIER 149

But from the whole mass of works of fiction published by Gautier,
the brief tales analyzed here, from la Morte amoureuse to Spirite,
emerge as structurally unique.

The credit is probably due to Gautier alone. The likeness of
some of the earlier stories to the tales Poe was producing at about
the same time in America! does not warrant inferences, it has been
observed, as to the influence of Poe. Equally fortuitous and no
less interesting is the parallel between the later products of Gautier
and the nineteenth-century German Novelle.

A purely casual remark of the author in Arria Marcella that he is
presenting ‘‘le simple récit d’une aventure bizarre et peu croyable,
quoique vraie, ? is nearly identical with Goethe’s famous definition:
“Was ist eine Novelle anders als eine sich ereignete unerhérte Begeben-
heit?’’? Aventure may be set off against Begebenheit, bizarre et peu
croyable against unerhôrte, vraie against sich ereignete, simple against
eine. Indeed, Gautier’s statement, put alongside of Goethe’s,
illuminates the latter, which, “the quintessential result of an investi-
gation . . . conducted intermittently for over thirty years,’ is
cast in somewhat oracular form. There is a meager possibility that
Gautier was acquainted with this definition. It is a fact that he
knew the Wahlverwandtschaften,> which includes the story of Die
wunderlichen Nachbarskinder, the first of Goethe’s narratives to be
presented to the public under the express title Novelle. But there
is not the requisite evidence to substantiate a belief that Gautier was
inspired by the German, and the remark which so neatly corre-
sponds to Goethe’s appears to be a felicitous accident.

1Tt may also be affirmed that there is a resemblance to Hawthorne in some of
Gautier’s work. In fact, the following criticism of Hawthorne, by Robert Louis Steven-
son (Essay on Hugo, p. 20), is applicable, in its entirety, to the author of Jettatura:
“There is a unity, an unwavering creative purpose, about some at least of Hawthorne’s
romances, that impresses itself on the most indifferent reader; and the very restrictions
and weaknesses of the man served perhaps to strengthen the vivid and single impression
of his works.’’ Gautier was acquainted with the fiction of Hawthorne (cf. Portraits con-
temporains, p. 157) and had been particularly impressed by the conception of a garden of
poisonous flowers in Rappacini’s Daughter, a story kindred in structure to Jettatura,
Avatar, ete. (cf. Histoire du Romantisme, p. 353; Fusains et eaux-fortes [Paris, 1907],
pp. 308-11).

2 Romans et Contes, p. 273. 3 Gespräche mit Eckermann, January 29, 1827.

4 Mitchell, ‘‘Goethe’s Theory of the Novelle,’’ Publications of the Modern Language
Association, XXX, 236. This article and Professor Mitchell's thoroughgoing study, Heyse
and his Predecessors in the Theory of the Novelle (Frankfurt, 1915), are the principal source
of the following remarks on the German definitions.

5 Histoire de l'art dramatique, I, 193; IV, 337.

661

150 Horatio E. SMITH

In practice, the epithet vraie is not equivalent to Goethe’s sich
ereignete. For obviously the adventure in Arria Marcella, or that in
Avatar, is not true, could not actually happen, in the same sense as
the events of Der Prokurator or of Ferdinand. Closer to Goethe in
this respect are Jettatura and Une Nuit de Cléopâtre, although these
sin against a principle enunciated by one of Goethe’s spokesmen:
“Tch leugne nicht, dass ich die Geschichten nicht liebe, die unsere
Einbildungskraft immer in fremde Lander nôtigen. Muss denn
alles in Italien und Sizilien, im Orient geschehen? Sind denn Neapel,
Palermo und Smyrna die einzigen Orte, wo etwas Interessantes
vorgehen kann?” As regards unity and novelty, however, Gautier’s
stories fully meet the German’s requirements. Likewise, Gautier’s
leisurely expositions find a counterpart in such stories as Die wunder-
lichen Nachbarskinder and Der Prokurator, his occasionally appended
codas a corresponding feature in Ferdinand.

Two other definitions, by Tieck and by Paul Heyse, stand out in
the ponderous mass of German criticism, and both of these have
their application here. Tieck, writing in 1829, insists with Goethe
upon unity and novelty and develops independently the theory that
the prime requisite is a single turning-point “von welchem aus sie
[die Geschichte] sich unerwartet véllig umkehrt, und doch natürlich,
dem Charakter und den Umstanden angemessen, die Folge ent-
wickelt. .... 2 Apparently Tieck would give his entire approval
to le Roi Candaule, where the action turns upon the discovery by the
queen that Gyges is inspecting her beauty, and to Une Nuit de
Cléopâtre, developed about the compact which Cleopatra makes with
the young Egyptian. In some of the other Gautier tales, such as
Avatar and Spirite, there is no prominent Wendepunkt, but perhaps
the theorist would be liberal in his interpretation, for he states that
each of the Novelas exemplares of Cervantes possesses such a turning-
point, yet investigation of these shows, I think, that it is frequently
not easy to determine which decisive act shall be so denominated.®

1Goethe, Werke, Weimarer Ausgabe, XVIII, 190, quoted by Mitchell, Heyse and
his Predecessors, p. 25.

2 Schriften (Berlin: Reimer, 1829), XI, p. Ixxxvi. The theory is discussed by
Mitchell, op. cit., pp. 33 ff.

2 Closest to the Novelle type are La Gitanilla and El Celoso Extremeño. In the
Coloquio de los Perros are one or two comments on problems of narration; cf. especially
where Cipion urges Berganza to refrain from digressions (p. 340 of the Leipzig, 1869,

662

Tue Brrer-NARRATIVE ART OF THEOPHILE GAUTIER 151

More specific and of special application to Gautier is the definition
made by Paul Heyse in order to set a standard for his Deutscher
Novellenschatz:

Im allgemeinen . . . . halten wir . . . , an der Regel fest der Novelle
den Vorzug zu geben, deren Grundmotiv sich am deutlichsten abrundet
und ....etwas Eigenartiges, Spezifisches schon in der blossen Anlage
verrät. Eine starke Silhouette—um . . . . einen Ausdruck der Malersprache
zu Hilfe zu nehmen—diirfte dem, was wir im eigentlichen Sinne Novelle
nennen, nicht fehlen, ja wir glauben, die Probe auf die Trefflichkeit eines
novellistischen Motivs werde in den meisten Fallen darin bestehen, ob der
Versuch gelingt, den Inhalt in wenige Zeilen zusammenzufassen, in der Weise,
wie die alten Italiener ihren Novellen kurze Uberschriften gaben, die dem
Kundigen schon im Keim den spezifischen Wert des Themas verraten.!

This is close to Gautier’s conception of brief-narrative unity as
exemplified in the series of stories under discussion,” the last of which
was published five years before the promulgation of Heyse’s theory
(1871). That is to say, here again are narrative methods parallel
and independent.®

Probably Gautier was not the only Frenchman to produce during
this period stories comparable in form to the Novelle. This structural
method would seem sufficiently elastic to lend itself—perhaps with
results not so lurid and of deeper literary import—to the purposes
of authors in power and doctrine quite different from him.

edition): ‘‘Quiero decir que la sigas [la novela] de golpe, sin que la hagas que parezca
pulpo, segun la vas afiadiendo colas.’’ Gautier refers several times to Cervantes’ stories,
but not critically (cf. Voyage en Espagne (Paris, 1899], pp. 118, 299; Guide au Louvre,
p. 274).

1 Deutscher Novellenschatz (München, 1871), Einleitung, pp. xix—xx.

2 Heyse quotes, for illustration of his theory, Boccaccio’s abstract of the ninth
Novella of the fifth day: ‘‘ Federigo degli Alberighi ama e non & amato; e in cortesia spen-
dendo si consuma, e rimangli un sol falcone, il quale, non avendo altro, da a mangiare alla
sua donna venutagli a casa: la qual cid sappiendo, mutata d’animo, il prende per marito
e fallo ricco.”’

Gautier’s stories may be summed up quite as succinctly. Cf. the abstracts given
in the course of this article. Cf. also the lines of Swinburne which epitomize Une Nuit de
Cléopâtre (Poems [London, 1904], III, 64):

And that great night of love more strange than this,

When she that made the whole world’s bale and bliss
Made king of all the world’s desire a slave,

And killed him in mid kingdom with a kiss.

3 Speculation about the reasons for these resemblances between Gautier’s work and
the short-story and the Novelle types remains idle until more is known about the French
field. Apparently the conditions of modern literature tend to develop two forms of brief
tale, as in English where we have not only the close-knit ‘‘short-story,’’ but the more
loosely woven—and clumsily named—‘‘long short-story.”

663

152 Horatio E. Smita

For Gautier, in any case, one point remains to be made. His
critics,! maintaining that he is above all a poet and a painter, like to
stress either the lyric or the pictorial element in his stories, and grant
him small skill in sheer narration. Without doubt their view that
fiction was of secondary interest to him is correct. Yet he accepted
its conditions more whole-heartedly than they think. There is a
suggestion of this truth in his letter to Sainte-Beuve: ‘Si j'avais
possédé la moindre fortune personnelle, je me serais livré unique-
ment à l'amour du vert laurier; mais dans la prose où je suis tombé,
j'ai toujours défendu les intérêts de l’art. ...”? And the evidence
proves that the structure of many of his tales is far from being either
haphazard or spontaneous, that the degree of unity he sometimes
attains is the result of deliberate plan, that after all, as a narrator, he
is no mean figure.

Horatio E. SMITH

Yate UNIVERSITY

1 Faguet, Sainte-Beuve, Lafcadio Hearn, et al.
2 Lovenjoul, op. cit., Introduction, p. xx.